BEACON: Cross-Domain Co-Training of Generative Robot Policies via Best-Effort Adaptation
BEACON: Cross-Domain Co-Training of Generative Robot Policies via Best-Effort Adaptation
This research introduces BEACON (Best-Effort Adaptation for Cross-Domain Co-Training), a framework designed to train generative robot policies using abundant source demonstrations and limited target demonstrations.
The paper proposes BEACON, a theory-guided framework for cross-domain co-training of generative robot policies that uses discrepancy-aware importance reweighting to improve robustness and data efficiency across sim-to-sim, sim-to-real, and multi-source manipulation settings without an explicit alignment objective.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
Source：arXiv Robotics — research abstracts · arxiv.org